Testimonial Smothering and Domestic Violence Disclosure in Clinical Contexts

Abstract

Domestic violence and abuse (DVA) are at last coming to be recognised as serious global public health problems. Nevertheless, many women with personal histories of DVA decline to disclose them to healthcare practitioners. In the health sciences, recent empirical work has identified many factors that impede DVA disclosure, known as barriers to disclosure. Drawing on recent work in social epistemology on testimonial silencing, we might wonder why so many people withhold their testimony and whether there is some kind of epistemic injustice afoot here. In this paper, I offer some philosophical reflections on DVA disclosure in clinical contexts and the associated barriers to disclosure. I argue that women with personal histories of DVA are vulnerable to a certain form of testimonial injustice in clinical contexts, namely, testimonial smothering, and that this may help to explain why they withhold that testimony. It is my contention that this can help explain the low rates of DVA disclosure by patients to healthcare practitioners.

1. Introduction

Domestic violence and abuse (DVA) are at last coming to be recognised as serious global public health problems. Nevertheless, many women with personal histories of DVA decline to disclose them to healthcare practitioners. Taking a naïve view, we might think this is surprising. Why would women with personal histories of DVA withhold this potentially crucial information from healthcare practitioners who might be able to help them? In the health sciences, recent empirical work has identified many factors that impede DVA disclosure, known as barriers to disclosure. In this paper, I offer some philosophical reflections on DVA disclosure in clinical contexts and the associated barriers to disclosure. Drawing on recent work in social epistemology on testimonial silencing (Fricker 2007; Dotson 2011), we might wonder why so many people withhold their testimony and whether there is some kind of epistemic injustice afoot here. I argue that ill persons with personal histories of DVA are vulnerable to certain forms of testimonial injustice in clinical contexts, namely, testimonial smothering (Dotson 2011), and that this may help to explain why they withhold that testimony. Testimonial smothering occurs when someone withholds or truncates their testimony, where the content of that testimony is such that misunderstanding is likely to have serious negative consequences, the audience has demonstrated a lack of competence in handling such testimony, and that lack of competence follows from pernicious, non-culpable ignorance on the part of the audience. It is my contention that this can help explain the low rates of DVA disclosure by patients to healthcare practitioners.

2. Domestic violence, disclosure, and screening

Domestic violence and abuse (DVA; also known as intimate partner violence) involve physical, sexual, psychological, and economic abuse against a person by an intimate partner (Heise et al. 1999). DVA are at long last coming to be recognised as serious global public health problems (World Health Organization 2013a). The vast majority of those who experience domestic violence and abuse are women. Globally, some 30% of women experience intimate partner violence (including sexual violence) at some point in their lives, and 38% of murdered women are murdered by their intimate partners (WHO 2013a: 2). Moreover, women who experience domestic violence and abuse are also more likely to suffer from other serious health problems, such as increased incidence of sexually transmitted infection, including HIV, alcohol (ab)use, mental health problems,¹ including depression and suicide, physical injury, and death from homicide (WHO 2013a: 21–30; see also Campbell et al. 2002). Note that these statistics do not necessarily include psychological and financial abuse.

It is not clear whether these statistics include trans people and in particular trans women. In a systematic literature review, Otero et al. (2015: 929) observe that the majority of studies on DVA in transgender, transsexual, and intersex couples suffer from a serious conceptual error, whereby relationships with at least one trans partner are categorised as homosexual relationships. Otero et al. (2015) found that the prevalence of DVA in transgender, transsexual, and intersex couples ranged from 18% to 80%. In a survey conducted in Scotland, 80% of trans people reported having experienced violent or abusive treatment from intimate partners, although only 60% recognised this behaviour as DVA (Roch et al. 2010: 6). In any case, there is little room for doubt that domestic violence and abuse are prevalent and dangerous.

It has been reported that most women experiencing DVA would not voluntarily disclose their history to physicians, and many would prefer for physicians to inquire about DVA (Friedman et al. 1992; McCauley et al. 1998; Elliot et al. 2002; Feder et al. 2006; Trevillion et al., 2014). Indeed, many women find that disclosure of their personal histories of domestic violence and abuse to healthcare practitioners is met with disbelief, victim blaming, and critical judgements (Flinck et al. 2005; Robinson 2010; Thapar-Björkert and Morgan 2010). As a result, there is a barrier between women with histories of DVA and the healthcare practitioners who treat them. These barriers have become known as barriers to disclosure.

Because of the high incidence rate of DVA and its negative health effects, removing these barriers – or at least minimising their consequences – is important. Some health researchers have recommended routinely screening patients for DVA (Campbell et al. 2002), although it is worth noting that there is no single agreed definition of routine screening (Waalen et al. 2000: fn. 1). There are numerous screening instruments for DVA, which typically involve short verbal questions or written questionnaires which directly elicit testimony from patients.² According to some approaches, every patient should be screened each time they make a medical visit. This is known as universal screening. According to other approaches, patients should only be screened when the practitioner suspects that they are a victim of DVA. Many prominent organisations have recommended the implementation of screening policies: the World Health Organization conditionally recommends such an approach (WHO 2013b).³ The spirit of these recommendations is captured in the words of the Emergency Nurses Association, who state that ‘the emergency nurse is an advocate for victims of domestic violence and has a duty to identify and report domestic violence’ (2003; quoted in Robinson 2010: 572).

The tendency of women with personal histories of DVA to withhold their testimony in clinical contexts gives rise to an interesting philosophical question. When a social group appears to be impeded from making testimony, and that impediment seems to be systematic or institutional in some way, we ought to consider whether there is some kind of injustice afoot. As I will explain in the following section, recent work in social epistemology deals with the question of epistemic injustice, that is, the matter of how people are wronged in their capacity as sources of knowledge (Fricker 2007).

3. Testimonial smothering⁴

We depend for the vast majority of our knowledge on the testimony of others. Whether we accept others’ testimony – whether we believe them, take their word for it, or ignore them – depends on our judgements about their credibility.⁵ The problem is that these judgements are very often guided by prejudicial identity stereotypes. Sometimes these prejudices strip people of the credibility they deserve, with the consequence that their testimony is unfairly rejected. Other times these prejudices grant people an excess of credibility, with the consequence that their testimony is believed when perhaps it should not be.⁶ Of course, both of these scenarios are undesirable for epistemic reasons. But there are also reasons to think that it is unfair, for example, to the speaker whose testimony is ignored because of their race or their gender. Fricker (2007) calls this testimonial injustice. In this essay, I argue that ill persons who experience domestic violence and abuse are especially vulnerable to a certain form of testimonial injustice, namely, testimonial smothering. Before taking this project any further, it is worth explaining exactly what is meant by this term.

In her critique of Fricker's view, Dotson (2011) identifies two practices of testimonial silencing, namely, quieting and smothering. Each of these terms are relevant to the topic of this paper, but since my focus here is on testimonial smothering and DVA disclosure, I will explain that concept in greater detail. A speaker is quieted when her testimony is undervalued or even completely ignored by its audience. When a person has been quieted, her audience does not recognise her as a good informant. Dotson writes:

The problem of testimonial quieting occurs when an audience fails to identify a speaker as a knower. A speaker needs an audience to identify, or at least recognize, her as a knower in order to offer testimony. (Dotson 2011: 242)

Among the principle causes of testimonial quieting is the phenomenon of negative epistemic stereotyping. This, according to Fricker (2007), is a form of identity prejudice. Consider how stereotypes undermine the testimony of women on matters which supposedly require intellectual rigour and cool-headedness. Testimonial injustice, according to Fricker, arises when someone's testimonial contribution is undervalued as a result of identity prejudice. For Fricker,

A negative identity-prejudicial stereotype is … [a] widely held disparaging association between a social group and one or more attributes, where this association embodies a generalization that displays some (typically, epistemically culpable) resistance to counter-evidence owing to an ethically bad affective investment. (Fricker 2007: 35)

Fricker proposes a model of conversational interaction which gives us a foothold in our effort to understand the nature of the phenomenon of testimonial injustice. In an ordinary (and, more to the point, successful) conversational interaction, the speaker and the hearer rely on stereotypes and heuristics to make judgements about the sincerity and reliability of their interlocutor. She remarks:

This model of the interaction between speaker and hearer helps us to see the mechanism whereby identity prejudice can distort a hearer's credibility judgement: it distorts the hearer's perception of the speaker. Applying the perceptual idiom to our chief example, we can say that the judgement of the jurors of Maycomb County is so distorted by prejudicial racial stereotype that they cannot, in that courtroom context, perceive Tom Robinson as anything but a lying Negro. Now in this example the jurors’ perceptions are shaped inter alia by prejudiced beliefs; the prejudicial racial stereotype determining their credibility judgements is in part doxastically mediated. (Fricker 2007: 36)

In the fictional case of Tom Robinson, his testimony was received but refused. His attempt to share his knowledge failed because his audience did not afford him sufficient credibility, as a result of their racial prejudices.⁷ Sometimes, however, a speaker does not get as far as sharing their knowledge at all, but rather, they withhold it. We withhold testimony for all sorts of reasons, for instance, to avoid offending someone or even simply to cut short a conversation that is taking too long to wind up on its own. Not all instances in which a speaker withholds her testimony are so innocuous. I will follow Dotson in calling the phenomenon of the coerced withholding of testimony ‘smothering’. And what is smothering? Dotson puts it thus:

Testimonial smothering, ultimately, is the truncating of one's own testimony in order to ensure that the testimony contains only content for which one's audience demonstrates testimonial competence. (Dotson 2011: 244)

In archetypal cases of testimonial smothering, a person who knows that p nevertheless withholds that testimony because they reasonably believe that their testimony will be refused outright or misunderstood in a way that leads the hearer to hold harmful beliefs.

Cases of testimonial smothering, according to Dotson, typically occur in the presence of three distinctive but related circumstances. These circumstances are as follows:

1) the content of the testimony must be unsafe and risky; 2) the audience must demonstrate testimonial incompetence with respect to the content of the testimony to the speaker; and 3) testimonial incompetence must follow from, or appear to follow from, pernicious ignorance. (Dotson 2011: 244)

It is worth taking a moment to reflect upon each of these circumstances and on how they relate to one another. According to Dotson, unsafe testimony ‘is testimony that an audience can easily fail to find fully intelligible’ (2011: 244). Risky testimony, for Dotson, ‘runs the risk of leading to the formation of false beliefs that can cause social, political, and/or material harm’ (2011: 244). This feature of Dotson's account of testimonial silencing is particularly helpful for understanding the barriers to disclosure faced by women with personal histories of domestic violence and abuse in clinical contexts.

The thought is that a piece of testimony is unsafe and risky for a given hearer if (i) the hearer could easily misunderstand the testimony, and (ii) if the hearer misunderstands it, then the resulting beliefs are likely to have harmful consequences. Dotson seems to employ a broad sense of the misunderstanding here. Dotson invokes Hornsby's reciprocity condition for successful linguistic exchange, according to which the hearer not only understands the speaker's words but also understands the speaker's words as they were intended to be understood by the speaker (Dotson 2011: 237–8; see also Hornsby 1995). As an example of unsafe and risky testimony, Dotson (2011: 244–5) gives the example of testimony about domestic violence in non-white communities. Referring to the work of Crenshaw (1991), Dotson explains how, when women of colour consider speaking about domestic violence in African American communities, they often withhold their testimony because of the risk that what they say will be taken to justify harmful stereotypes about African Americans:

Some, though certainly not all, African Americans have considered the ramifications of testimony about certain kinds of occurrences, like domestic violence and/or rape, to be a detriment to African American communities at large, often at the expense of those who suffer from domestic violence and/or rape.⁸ (Dotson: 2011: 245)

The thought is that the content of the testimony is unsafe because audiences are likely to misunderstand it, incorrectly taking reports of discrete instances of wrongdoing as evidence of more general behavioural trends and thereby failing to understand what the speaker wished to communicate with her words. There is also a risk that people who make such testimony will be treated as ‘spokespersons’ for their community, which may bring with it an increased risk of epistemic injustice (Davis 2016).⁹ On Dotson's view, the content of the testimony is risky because, if the hearers of the testimony misunderstand it, they are likely to form damaging, false beliefs about African American people. It seems to me that it may be possible to expand this condition for testimonial smothering to include not only the hearer's doxastic responses but also their behavioural responses. I will expand on the relevance of this to DVA disclosure in clinical contexts later.

The second circumstance of testimonial smothering, on Dotson's account (2011: 245), is that the hearer demonstrates to the speaker that she is an incompetent recipient of testimony. Testimonial competence on the side of the hearer involves some degree of proficiency in at least the following two skills: on the one hand, your testimonial competence depends on your ability to understand what you are told, and on the other hand it depends on your ability to recognise when you do not (or you are not likely to) understand what you are told. Consider the experience of listening to a lecture on theoretical physics as a non-expert (Dotson 2011: 245). The competent non-expert can not only understand some of what she is told in the lecture, but also recognise when she does not understand the theoretical physicist's testimony. How does this link to testimonial smothering?

Well, testimonial smothering requires that the speaker withholds or truncates her testimony because the hearer has demonstrated that she is ill-equipped to interpret the speaker's testimony correctly or to appreciate her own limitations. By including this condition, Dotson anticipates potential counter-examples in which someone withholds their risky and unsafe testimony because they have unreasonable doubts about the competence of the hearer. The thought is that testimonial injustice does not occur when someone withholds their testimony from someone who has nevertheless demonstrated that they are a competent recipient of testimony. On Dotson's view, when a speaker's testimony is smothered, it is because the hearer has demonstrated to the speaker that they are (or would be) an incompetent recipient of the speaker's testimony (providing the other conditions are satisfied).

As it happens, there are good reasons to think that this way of framing the condition is too restrictive. Certainly, a speaker's testimony might be smothered if their hearer is demonstrably incompetent. However, there are circumstances in which a speaker's testimony may be smothered if the hearer fails to demonstrate that they are a competent recipient of the speaker's testimony. As Beecroft explains (Ms: 7–9), speakers can come to expect incompetence on the part of their hearers quite reasonably, especially against the background of unjust cultural discourses, or in response to the social identity of the hearer.¹⁰

A third and final circumstance of testimonial smothering is that the testimonial incompetence of the hearer must result from (or appear to follow from) pernicious, situated ignorance. What, then, is this kind of ignorance? Someone in a state of pernicious, situated ignorance lacks knowledge as a result of their social positioning. Dotson describes it thus:

Situated ignorance, which follows from one's social positioning, is a result of epistemic limitation that fosters a kind of epistemic distance between those not in possession of that limitation and those who do possess the limitation. (Dotson 2011: 248)

Epistemic distance between two or more people exists when there is a gap between their respective worldviews, such that they have different ways of seeing and understanding the world. Differences in race, gender, and social and economic status can all contribute to the growth of epistemic distance between persons or peoples (Dotson 2011: 248). According to Dotson's account, testimonial smothering is likely to occur if the failure of the hearer to demonstrate testimonial competence results from pernicious, situated ignorance.

This brief summary is sufficient to explain what Dotson means by the practices of silencing, and to demonstrate how her insight can be applied to some cases. Testimony can be quieted or smothered, depending on whether the hearer fails to afford the speaker their due credibility, or the speaker withholds her own testimony because she reasonably believes that her testimony will be misunderstood in a way that leads her hearer to form harmful beliefs. Put this way, smothering occurs as a result of anticipated quieting. The speaker withholds some or all of her testimony because she believes that testifying will not serve its intended purpose. In anticipation of the predicted failure of her attempt to convey her knowledge, the speaker refrains from speaking at all.

To recap: the thesis of this paper is that women with personal histories of domestic violence and abuse are especially vulnerable to testimonial smothering in clinical contexts. Before introducing the DVA-related reasons that some patients smother their testimony, it is important to note that ill persons are already at a heightened risk of epistemic injustice in healthcare settings. Ill persons with histories of DVA are thus socio-epistemically disadvantaged prior to the introduction of additional, DVA-related factors that further aggravate the epistemic injustice.

4. Epistemic injustice and illness

In some recent philosophical work it has been persuasively argued that ill persons are especially vulnerable to epistemic injustice (Carel and Kidd 2014; Kidd and Carel 2017). This is the first barrier to disclosure of DVA in clinical contexts. Before we focus on the specific problems associated with DVA disclosure, we must consider how stereotypes and prejudices about illness already hinder patients’ ability to share their knowledge via testimony.

The two most widely discussed varieties of epistemic injustice are testimonial injustice and hermeneutical injustice. Since the subject of this paper is testimonial injustice, I will set the topic of hermeneutical injustice in healthcare to one side for now.¹¹ There are, as Carel and Kidd explain, ‘several ascending ways’ in which ill persons suffer testimonial injustice (2014: 531–2). The testimony of ill persons may simply be ignored by healthcare practitioners. Their testimony may not be ignored but nevertheless it may be excluded from consideration, either because it is judged to be uninformative or because its presentation precludes it from serious consideration. Alternatively, the ill person's testimony may be subordinated to the (epistemic) authority of the healthcare practitioner. These are all forms of testimonial injustice.

Just as all sorts of stereotypes lead to the prejudices that guide our judgements about people's testimony, so stereotypes about ill persons lead to prejudices that guide healthcare practitioners’ judgements about the credibility of their patients. Some of these stereotypes have to do with the manner in which the testimony is presented. Ill persons’ testimonies may be emotionally charged, and understandably so, but such emotional displays may be taken as evidence that a speaker lacks credibility. This is perhaps because it suggests that they are thinking irrationally. Ill persons’ testimonies may also include medically irrelevant information. The inclusion of irrelevant information in testimony may also be treated by the hearer as evidence that a speaker lacks credibility. This could be because the inclusion of irrelevant information suggests that a speaker either lacks the relevant medical competence or that their intention is not to communicate what they know, but rather to elicit a particular response from their hearer.¹²

Blease et al. (2017) have argued that patients with chronic fatigue syndrome (CFS; also known as ME) suffer both testimonial and hermeneutical injustice in clinical contexts.¹³ They offer evidence from practitioners and patients in support of this claim. They point to recent surveys which indicate that in the UK and in Australia there is significant scepticism among general practitioners about the legitimacy of CFS (Blease et al. 2017: 553). If a practitioner harbours unwarranted doubts about the legitimacy of a patient's illness – let alone whether it is correctly diagnosed in this case – this could lead to further scepticism about the credibility of CFS patients’ testimonies.¹⁴

Another relevant example of increased vulnerability to epistemic injustice in clinical contexts can be found in the case of pregnant women in clinical contexts. Freeman argues that pregnant women experience epistemic injustice when the

claims that [they] make about their bodies are not taken seriously; when their desires and requests are systematically undermined, overlooked, or ignored; when their agency fails to be recognized; and when, as a result of being unheard and ignored, they are demoted to occupy a position of powerlessness. (Freeman 2015: 44–5)

Pregnant women's testimony is not given sufficient evidential weight in medical practitioners’ deliberations as a consequence of two factors, Freeman argues (2015: 45). The first of these factors is the pervasiveness of ‘visual paradigms of knowledge’ regarding pregnancy, according to which physicians’ observations are the best source of medically relevant information about pregnant women and their bodies; whereas those patients’ privileged epistemic access to their own bodily sensations is marginalised. The second is the widespread reliance by practitioners on medical technologies that fit within that visual paradigm, such as ultrasound imaging. Consequently, for example, evidence that can be gained via such medical technologies may wrongly be given greater evidential value than the evidence provided by women's own testimony. Of course, it is not always the case in medical contexts that patients have privileged epistemic access to the relevant information. However, when they do have such epistemic privilege but are nevertheless given diminished credibility by the medical practitioners that are attending them, they suffer epistemic injustice.

Psychiatric patients may be especially vulnerable to epistemic injustice in clinical contexts, according to Crichton et al. (2015), who argue that both global and specific factors contribute to the epistemic injustice faced by psychiatric patients. They identify the following global factors: (i) the social, economic, and cognitive consequences of mental disorder itself; (ii) the prioritisation of non-testimonial evidence by healthcare practitioners; and (iii) pernicious stereotypes about mental illness (Crichton et al. 201: 67–8). The specific factors that contribute to epistemic injustice in psychiatry are associated with particular conditions. For example, Alzheimer patients’ credibility is damaged by the often false belief that Alzheimer's disease causes ‘global and severe’ cognitive impairment (Crichton et al. 2015: 68–9). Thus both global and specific factors lead practitioners to deny psychiatric patients the credibility they deserve, leading to unjust responses to their testimony (Crichton et al. 2015: 65).

There is ample evidence that ill persons are vulnerable to epistemic injustice in healthcare settings. I have focused here on mentioning the ways in which ill persons are vulnerable to testimonial injustice in clinical contexts. In the following section of this essay, I will present what I take to be compelling evidence that ill persons with personal histories of DVA are especially vulnerable to testimonial injustice.

5.1. Testimonial smothering and (barriers to) DVA disclosure

In theory, quieting and smothering are closely related: a speaker's testimony is smothered when they withhold their testimony because they reasonably believe that their testimony will be misunderstood in a way that leads the hearer to form harmful beliefs. We will see that in practice this is also the case. In the following sections, I will present evidence from patients and evidence from practitioners. It is important at this point about what exactly would count as confirmatory evidence for this hypothesis. Remember, Dotson does not provide necessary or sufficient conditions for testimonial smothering; rather, she describes three circumstances which systematically accompany instances of testimonial smothering. These circumstances are closely connected with thwarted DVA disclosure in clinical contexts, or so I will argue in the following paragraphs. This is, therefore, an inference to the best explanation. The best explanation of the phenomenon that the majority of women do not disclose personal histories of DVA to healthcare practitioners in clinical contexts is that their testimony is smothered. In support of this claim, I will draw on empirical evidence about why women with personal histories of DVA withhold or truncate their testimony in clinical contexts. (It is important to note that this evidence is primarily drawn from the USA and the UK. Consequently, I would caution against generalising these results uncritically.) First, I argue that DVA disclosure is a kind of ‘unsafe or risky’ testimony. Second, I present some evidence for the claim that women with personal histories of DVA lack evidence that their audience is testimonial competent (in a context where such competence cannot be taken for granted) and often accompanied by situated ignorance about DVA.¹⁵

5.2. DVA disclosure as unsafe and risky testimony

Women with personal histories of DVA often withhold or truncate their testimony in clinical contexts. It has been estimated that approximately two-thirds of women with personal histories of DVA do not disclose them in clinical contexts (Friedman et al. 1992; McCauley et al. 1998). More recent findings suggest that only 5% of patients with histories of DVA are identified by healthcare practitioners (McGarry and Nairn 2015), at least in part because those patients do not disclose them. Why, then, do some women with personal histories of DVA not disclose them to healthcare practitioners? There is considerable evidence that DVA disclosure is a kind of unsafe and risky testimony. According to Dotson, as explained above, testimony is unsafe if it can easily be misunderstood by the hearer. Testimony is risky if misunderstanding it can lead to ‘social, political, and/or material harm’ (Dotson 2011: 244). If DVA disclosure is indeed a kind of unsafe testimony, we would expect to find evidence that it is easily misunderstood. Indeed, evidence from patients and practitioners alike supports this claim.

The evidence discussed in the following paragraphs will show that it is reasonable to suppose that DVA disclosure can easily be misunderstood and that its misunderstanding can have serious consequences for patients. To satisfy the first condition of Dotson's account of testimonial smothering, it only remains to show that DVA disclosure is risky and unsafe. We can find evidence of this in the fears reported by both patients and practitioners. McCauley et al. (1998) found that shame prevented women from disclosing their personal histories of DVA to healthcare practitioners. One woman interviewed in their study said:

[G]oing to a hospital for domestic violence is like going to the sexually transmitted disease clinic … you feel like the doctors look at you like you're dirty or you weren't protecting yourself. (McCauley et al. 1998: 552)

Sexually transmitted infections are widely stigmatised and often prove to be a significant source of shame for many people, sometimes to such an extent that they prevent people from taking precautions or seeking medical treatment.¹⁶ How then do shame and stigma lead women with personal histories of DVA to anticipate that disclosing their personal histories to their healthcare practitioners will lead those practitioners to form harmful beliefs? For a plausible explanation, let us briefly consider the nature of shame and the related attitude of contempt. We form these attitudes when we judge that someone has failed to live up to standards (Mason 2003, 2010). Contempt, Mason writes, ‘present[s] its object as low in the sense of ranking low in worth as a person in virtue of falling short of some legitimate interpersonal ideal of the person (Mason 2003: 241). The same, I take it, may be said for shame: Mason writes that ‘to experience shame is to experience oneself as diminished in merited esteem on the ground that one has violated some legitimate ideal of character’ (2010: 417–18). Shame and contempt are thus closely related. While shame is what we feel towards ourselves when we fail to live up to standards, contempt is what we feel towards other people when they fail to live up to standards. Since shame and contempt both (ideally) respond to evidence of failing or deficiency, it is plausible that a person who feels ashamed may also believe that they deserve other people's contempt. Now, by equating domestic violence and sexually transmitted disease, the woman interviewed suggests how feelings of shame and stigmatisation associated with domestic violence may prevent women from seeking support, including disclosing DVA to healthcare practitioners. Moreover, it is reasonable to suppose that a patient who is ashamed of their personal history of DVA may fear that their testimony will provoke the contempt of the healthcare practitioner who attends them, where contempt arises from the mistaken judgement that the patient's victimhood is a consequence of their failure to live up to standards.¹⁷

Related concerns about social class also prevent women from disclosing personal histories of DVA to healthcare practitioners. Indeed, McCauley et al. (1998: 552) note that women of all socio-economic backgrounds tended to associate domestic violence with lower socio-economic backgrounds than their own. In other words, most people thought domestic violence was ‘beneath them’. One woman interviewed by McCauley et al. said, ‘When I first came here, I thought it would be a group of low-lifes … You know, people that break beer bottles, scream and argue on Saturday nights’ (McCauley et al. 1998: 522). The notion that domestic violence is a problem for members of supposedly ‘lower’ social classes appears to cause women who experience DVA to feel ashamed of their own histories of DVA, regardless of their social background. Shame and related attitudes thus present another barrier to DVA disclosure.

To be clear, what I have identified is probably best understood as a sympathetic extension of the view of testimonial smothering proposed by Dotson. Where Dotson focuses on harmful beliefs, I have suggested that other complex attitudes can contribute to testimonial smothering. In particular, I have suggested that it is possible that some women with personal histories of DVA withhold their testimony because they believe that it will provoke the contempt of the healthcare practitioner.

Patients’ fears are widely recognised in the empirical literature as one of the barriers to DVA disclosure in healthcare settings (McCauley et al. 1998). These fears reveal some of the reasons why women with personal histories of DVA may withhold or truncate their DVA testimony in those (and other) contexts. These fears can be divided into general fears and specific fears. General fears are fears about DVA disclosure that apply to most women; specific fears are fears about DVA that are relevant for groups of women with particular characteristics such as age, social class, race, and immigration status.

I will first describe some of the general fears that are felt by patients towards DVA disclosure. Among the women surveyed by McCauley et al., women frequently mentioned that they worried that if they disclosed a personal history of DVA to a healthcare practitioner, the practitioner would blame them for the abuse they have suffered at the hands of their abusers (McCauley et al. 1998: 553). Some women fear that if they disclose their personal histories of DVA to their healthcare practitioner, this will be passed on to the abusive party. Indeed, a third of the women interviewed by McCauley et al. (1998) indicated that they were prevented from disclosing their history of DVA to healthcare practitioners by fear of repercussions from their abusers. Reflecting on what would happen if she reported her abuse and her abuser found out, one woman commented: ‘I'll get beat up more. I mean, God forbid they're gonna give him something to be angry about. He's angry about nothing’ (McCauley et al. 1998: 553). This is just an introduction to the dangers that are posed by DVA disclosure.

Now I will describe some more specific fears that prevent different groups of women from disclosing personal histories of DVA to healthcare practitioners in clinical contexts. These fears reveal both the risks incurred in disclosing DVA and the possibility of that disclosure leading to misunderstanding on the part of its recipient. First, women with children were prevented from disclosing DVA to their healthcare practitioners by concerns that doing so would have harmful consequences for their children: ‘They want to check your children for sexual abuse … I don't want my children up on nobody's table at the age they are with their legs spread’ (McCauley et al. 2018: 552). Second, immigration status also appears to be a barrier to DVA disclosure. The fear that DVA disclosure could lead to deportation prevents some immigrant women in the US Midwest from accessing outreach support organisations for DVA victims (Reina et al. 2014). This perceived risk prevents some women from disclosing their personal histories of DVA to healthcare practitioners. Third, race presents barriers to DVA disclosure too. Indeed, this is the very example that Dotson cites (2011: 245). To reiterate Dotson's point: many Black women refrain from disclosing personal histories of DVA because they are concerned that such testimony might be misunderstood in ways that bring about harmful consequences for Black people. Fourth, age appears to be a barrier to DVA disclosure: generational differences in norms regarding ‘privacy about domestic affairs’ may prevent older women from disclosing DVA to healthcare practitioners (Zink et al. 2004). Older generations, it is supposed, tend to place special value on maintaining the privacy of one's personal life. Sharing intimate details about one's private life, including (or, one imagines, especially) information about experiences of violent or abusive behaviour, is perceived to come at heightened social cost. So, in addition to the risks faced by all (or most) women who disclose personal histories of DVA, the specific fears described here indicate that other aspects of women's social identities may add further risks to DVA disclosure in clinical contexts. All of these concerns may contribute to the smothering of those women's testimony. The analysis here can be sharpened by highlighting a link between patients’ fears and what McKinney calls ‘unjust locutionary extraction’ (2016). McKinney explains that, while our speech is ideally a manifestation of our agency, our words can also be ‘used against us’ (2016: 259–60). One way in which a speaker can undergo unjust locutionary extraction is when they are made to make an utterance which licenses wrongs against them (McKinney 2016: 265). Given the patient fears enumerated in the preceding paragraph, we might think that when women withhold personal histories of DVA, they do so because they do not want their speech to be used against them. In particular, they may be concerned that disclosing DVA will unleash a cascade of negative consequences. We will return to this thought later. One way, then, to understand what is going on here, is that concerns about the risk of unjust locutionary extraction contribute to the testimonial smothering of women with personal histories of DVA.¹⁸

So far I have focused on patients’ fears about DVA disclosure. Practitioners’ fears about the risks associated with DVA disclosure are also informative. Healthcare practitioners are in a position of relative authority in the sense that, to some degree, how they respond to the patient's testimony can have major consequences for them. However, the institutional setting that grants practitioners their authority also truncates it in pernicious ways. It seems that practitioners have relatively little autonomy concerning how, or indeed, whether, their patients’ DVA disclosures will be followed up. Moreover, in recent work on the moral dispositions and agency of frontline public service providers in bureaucratic institutional settings, Zacka suggests that workers often adopt pathological reductive dispositions towards service users (2017) as a way of coping with the cognitive dissonance that arises from the difficulties involved in working with limited resources whilst remaining responsive to competing demands, among other factors (Zacka 2017: 123). Zacka identifies three such dispositions: indifference, enforcement, and caregiving. I will sketch these very roughly here. Indifference leads to withdrawal from the messiness of individual cases and allows workers to function as morally disengaged ‘people processors’ (Zacka 2017: 101–4). Enforcers and caregivers are both much more engaged with their institutional roles, but they have starkly contrasting priorities. Enforcers focus on and prioritise the upholding of rules and regulations to protect their organisations from abuse by service users, perhaps at the expense of some deserving service users (Zacka 2017: 106–9). Caregivers, by contrast, focus on and prioritise the particular circumstances and needs of their clients, perhaps at the expense of efficiency (Zacka 2017: 104–6). Which of these dispositions a service provider adopts makes a considerable difference to how they operate within the margin of discretion afforded to them in their institutional roles (Zacka 2017).¹⁹

It seems plausible that the pressures associated with working in clinical settings can cause practitioners to adopt any of a variety of pathological dispositions towards their patients. These dispositions may shape how healthcare practitioners interact with patients and how they exercise their discretion. In particular, adopting certain dispositions in their roles may cause some practitioners to refrain from screening patients for DVA. Strikingly, it has been reported that some practitioners fear that screening their patients for DVA runs the risk of ‘opening Pandora's box’, that is, that so doing will set in motion a chain of negative consequences for the patient (Sugg and Inui 1992; McCauley et al. 1998). This fear is compounded by concerns about the lack of resources to follow up DVA disclosure. Indeed, the lack of resources to adequately follow up DVA disclosure was reported as a barrier to DVA screening by 63.6% of the articles reviewed by Sprague et al. (2012: 596). If practitioners’ fears about following up on DVA disclosure cause them to withdraw from their patients and refrain from screening their patients or to avoid the topic (perhaps unconsciously), then an important opportunity for patients to disclose personal histories of DVA will be lost. Moreover, such reluctance may be interpreted by patients as a sign that they should not disclose, further contributing to testimonial smothering. This lends support to the claim in the previous paragraph: it seems that patients’ fears that their speech could be used against them are, at times, well-founded. These examples of practitioners’ fears may be evidence of pathological indifferent or caregiving dispositions. But frustration can cause practitioners to form more hostile dispositions too. Indeed, various frustrations, including frustrations about the lack of resources to help women who disclose personal histories of DVA, led one of the nurses interviewed by Robinson (2010) to comment: ‘I don't have time to hear a 30-minute story about it. You're a grown person: get out of it. That's horrible, I shouldn't be saying that’ (2010: 574).²⁰

It is reasonable, then, to conclude that DVA disclosure is a kind of unsafe and risky testimony. Testimony that discloses a personal history of DVA is easily misunderstood and carries with it the risk of serious negative consequences for the patient. When women refrain from making a DVA disclosure in clinical contexts because of the perceived risks involved in making such a disclosure, it is plausible that their testimony has been smothered.

5.3. Testimonial incompetence, pernicious ignorance and DVA disclosure

There is considerable evidence that many women who withhold or truncate their testimony, do so because of concerns about the testimonial incompetence of their hearers in clinical contexts, that is to say, of their attending practitioners. Often, this incompetence is due to pernicious ignorance. Owing to the interrelatedness of these factors, I present them together. Beecroft's amendment to Dotson's account of testimonial smothering is relevant here (Ms). The thought is that there are circumstances in which a would-be speaker can have their testimony smothered simply because their audience fails to demonstrate that they possess the relevant competencies. Indeed, where the speaker's testimony is unsafe and risky, it seems reasonable for a speaker to withhold or truncate their testimony until they receive some appropriate kind of assurance that the hearer is a competent recipient of their testimony.

DVA disclosure seems to fit this description well. I have argued that DVA disclosure is a kind of unsafe and risky testimony; unsafe, in that it is easily misunderstood, and risky, in that those misunderstandings can yield serious negative consequences. The influence of stereotypes about DVA, its victims, and its perpetrators, all stack the deck against women's testimony. It is plausible, I would suggest, that against this background, women would not be unreasonable in supposing that their healthcare practitioners lacked the relevant testimonial competence, at least unless they possess additional evidence to the contrary. Interestingly, it has been shown that women with histories of DVA are much more likely to disclose their experiences if there are some visible indications that their healthcare practitioners are minimally receptive to such testimony. For example, the presence of posters or brochures was enough to reassure half of the women surveyed by McCauley et al. that they could disclose their personal histories of DVA (McCauley et al. 1998: 553). This may suggest that it is assumed that practitioners will not welcome DVA disclosure, but that this assumption can be overridden by evidence of receptiveness and competence on the part of healthcare practitioners.

How do testimonial incompetence and pernicious ignorance present barriers to disclosure in clinical contexts? In a systematic review of the literature on barriers to DVA screening, Sprague et al. (2012) identify a diverse range of factors which prevent healthcare practitioners from screening patients for DVA. It is notable that over two-thirds (68.2%) of the literature reviewed by Sprague et al. reports that factors such as a ‘lack of knowledge, education, or training regarding screening’ present significant barriers to DVA screening (2012: 596). We might interpret these finds as evidence that a significant proportion of healthcare practitioners refrain from screening their patients for DVA because they lack the relevant competencies or because they lack the relevant knowledge for dealing with DVA disclosure.

Women who do not fit certain stereotypes are at an epistemic disadvantage when it comes to disclosing personal histories of DVA. This thought is captured well by the criminological concept of the ‘ideal victim’. According to Christie, ‘ideal victims’ are people who ‘are most readily given the complete and legitimate status of being a victim’ when they are affected by crime (Christie 2018: 12). Christie outlines a number of attributes which are typically held by ‘ideal victims’ (2018: 12–13). Ideal victims are weak, they are engaged in respectable activities, and they are found in locations where they cannot be blamed for being. Think, for example, of an elderly person doing their shopping at the local greengrocers. Ideal-victim status also depends on the attributes of the perpetrator. For the victim to be an ideal victim, the perpetrator ought to be stronger than the victim and, significantly for this project, personally unknown to them. Non-ideal victims of crime are not given victim status. This can have the consequence that they are not provided access to appropriate support services. Presumably, it can also mean that when they report what happened to them, their testimony is not believed, either because they suffer a deficit of credibility, or because the perpetrator is granted a credibility-excess.²¹ Thinking about how the construction of the ideal victim and the corresponding notion of the ideal perpetrator manifest themselves in relation to DVA may help us to understand the barriers to disclosure in clinical contexts.

We see the concept of the ideal victim at work in nurses’ decisions whether to screen a patient for DVA. These decisions are sometimes guided by judgements about the demeanour of both the patients and the person(s) who accompany them to their appointments (Robinson 2010). One nurse interviewed in Robinson's study commented: ‘You have to look at the patient's demeanor and the demeanor of the person that is with them. Do they (the person accompanying the patient) have a dominating personality and is the patient shy and quiet?’ (2010: 574). This judgement, that victims of DVA are submissive and abusers are dominating, and the accompanying heuristic, seem to be based on the pervasive social constructions of the ideal victim and the corresponding ideal perpetrator. Women who match the profile of the ideal victim, or whose abusers match the profile of the ideal perpetrator may be detected and screened as a result of this heuristic. Nevertheless, this is concerning. It may be the case that many cases of abuse will fit this stereotype, but there is more than one way in which heuristics such as this one can mislead us. Relying on it as a way of determining whether a patient is at risk of DVA carries the significant risk of overlooking victims who do not fit the image of the ideal victim, upon which the heuristic is based.

Perceptions of social class have also been shown to guide healthcare practitioners’ judgements about the likelihood that a patient is a victim of DVA. In a study of 275 nurses conducted by Moore et al. (1998) 92% of respondents stated that women with middle- and upper-class social backgrounds were unlikely to experience DVA. This is another example of a barrier to disclosure that results from the social constructions of the ideal victim and the ideal perpetrator. Perhaps the thought is that people, in particular men, from middle- and upper-class social backgrounds are less likely to treat their intimate partners violently or abusively. In short, it may be the problem is not so much that middle- and upper-class women are non-ideal victims, but rather, that middle- and upper-class men are non-ideal perpetrators. Admittedly, this reading is speculative, but if it is correct, then it is evidence of another barrier to disclosure presented by pernicious ignorance.

This is evidence of situated ignorance on the part of healthcare practitioners. Of course, for it to be the case that these instances of situated ignorance contribute to the testimonial smothering of patients with personal histories of DVA, it must also be the case that those patients believe that they will not be recognised as genuine victims because they do not fit the stereotype of the ideal victim. Admittedly, this is harder to show, although it is plausible that some of the various patients’ fears mentioned above (e.g. especially fears regarding victim-blaming and other negative affective responses) are evidence that patients suspect that they will be stereotyped.

However, not all stereotypes about DVA can be understood as consequences of the social construction of the ideal victim. For instance, the influence of stereotypes about pregnancy and domestic violence also seem to lead to testimonial smothering. Due to the widespread but false belief that pregnancy can offer women a respite from abuse, domestic violence in pregnancy has been described as ‘the silent enemy’ (Scobie and McGuire 1999: 259). This stereotype may prevent healthcare practitioners from screening for DVA. Given that pregnant women already face a heightened risk of testimonial injustice in healthcare contexts (Freeman 2015), it is not unreasonable to suppose that this is another example of situated ignorance about domestic violence that contributes to the smothering of women's testimonies in clinical contexts.

There is considerable evidence that women who do not disclose their personal histories of DVA to healthcare practitioners in clinical contexts are victims of the epistemic injustice of testimonial smothering. That being said, there are other possible explanations that ought not to be overlooked. In particular, there is empirical evidence that some women's nondisclosure of personal histories of DVA is the result of a lack of understanding on their part; in particular, some women do not recognise their abusers’ abusive behaviour as abuse (Francis et al. 2017). We might understand this as a form of hermeneutical injustice. It is also important to note that not all women's decisions not to disclose personal histories of DVA are the consequence of epistemic injustice. Nevertheless, it is clear that the testimony of many women with personal histories of DVA is smothered in clinical contexts.

6. Conclusion

This essay began by asking why women with personal histories of domestic violence and abuse would withhold this information from healthcare practitioners. I have argued that there are good reasons to believe that many women withhold this testimony as a result of testimonial smothering. Evidence about barriers to disclosure lends credence to the claim that DVA disclosure is a kind of unsafe and risky testimony, and that its recipients are often epistemically incompetent, often owing to pernicious, non-culpable ignorance. Women with personal histories of DVA must overcome a variety of barriers before they can disclose them to healthcare practitioners. Some of these barriers are formed from the epistemic injustices faced by many ill persons in clinical contexts, but stereotypes about domestic violence, its victims, and its perpetrators give rise to yet more. In enumerating these barriers, this essay has offered a philosophical explanation of the barriers to DVA disclosure in clinical contexts. Nevertheless, I have stopped short of prescribing a solution. I hope that by analysing the problem of non-disclosure as a function of testimonial smothering, this essay may help us to avoid interpreting non-disclosure as pathological or irrational behaviour on the part of the victims. This may help us to steer clear of paternalistic responses to non-disclosure which only serve to damage the interests of DVA victims. There is much to be done.²²